The influenza virus hemagglutinin (HA) induces both binding and low pH- triggered fusion of a viral envelope to endosomal membrane resulting in release of viral genome into cytoplasm. A stable hemifusion intermediate without full fusion is induced by the HA ectodomain linked to the membrane via a lipid anchor, suggesting that wild type HA ectodomain is also capable of inducing hemifusion while the transmembrane, and, perhaps, cytoplasmic domains of HA are required for the transition from hemifusion to full fusion. To test this hypothesis, HAs in which transmembrane or cytoplasmic domains were truncated or substituted by unrelated sequences from either fusion proteins of other enveloped viruses or from non-viral integral membrane proteins will be expressed in cells. These HA-expressing cells will be bound to red blood cells or planar lipid bilayers and their fusion activity investigated by fluorescence imaging and time-resolved electrical admittance measurements to detect small fusion pore formation and its subsequent growth. The influence of HA structure on the enlargement of fusion pores will be studied to address whether the fusion protein controls this stage of fusion. Particular emphasis will be placed on detecting the hemifusion intermediate by using fluorescence video-microscopy and monitoring the redistribution of membrane probe between the external leaflets of fusing membranes. The redistribution of the membrane probe prior to the opening of small fusion pore will report the formation of a hemifusion intermediate. These studies will lead to a better understanding of the function of distinct topological domains of HA and will elucidate intermediates of membrane fusion, which is of practical importance for controlling the early stages of viral infection.